23 



three secondary radials, beyond which our sp8cimen is not pre- 

 served. If the last secondary and tertiary radials are axillary, 

 there are six arms in this series. By this arrangement, if the 

 arms are uniform, there are thirty-eight arms to this species. But, 

 we cannot be certain that all the arms are double, as we know in 

 some instances in other species such is not the case. The greater 

 probability is, however, that they are uniform and that the species 

 possesses thirty-eight arms. The ambulacral openings come to- 

 gether on the first tertiary radial,' and hence, if the projecting 

 radial series were broken away, at this place, there would appear 

 to be only nineteen ambulacral openings to the vault. There are 

 no inter-secondary or inter- tertiary radials. 



The first regular interradial is smaller than a second primary 

 radial. It is followed by two smaller plates in the second range, 

 and these by three smaller plates in the third range, which sepa- 

 rate the lower part of the first secondary radials. Above this 

 point there is some difference in the areas, and the plates are 

 small and thrown out of any definite arrangement by being con- 

 tinued up the sides of the projecting arm series as well as cov- 

 ering the depressed interradial area and graduating into the 

 plates of the vault without any line separating the plates of the 

 calyx from those of the vault. In one of the areas there are in 

 the fourth irregular ranga four plates and in another area there 

 are five, these separate the first tertiary radials. The first plate 

 in the azyzons area is smaller than a first primary radial and in 

 line with them. It is followed by two plates, in the second range, 

 each of which is nearly as large as a first regular interradial, and 

 these are followed by a range of three plates which are some- 

 what smaller and separate the third primary radials. There are 

 four plates in the fourth range and above this range the plates, 

 in our specimen, are covered with the matrix so as to obscure 

 the sutures, but the area above this does not seem to be differ- 

 ent from the regular areas. 



The vault is convex and covered with large polygonal plates 

 over the ambulacral areas, each one of which is possessed of a 

 large ventricose central node, while the interradial areas are 

 abruptly sunken and covered with small plane plates. The pro- 

 boscis appears to have been quite small and central though 

 broken away from our specimen. The ventricose vault plates in 

 this species are much like they are in Actinocrinns moniicul 'it- 

 ems and the swelling may be due more or less to silicification. 



